# Qwen3.5 397B vs Qwen3.6 Plus

> Side-by-side benchmark comparison for Qwen3.5 397B and Qwen3.6 Plus across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/qwen3-5-397b-vs-qwen3-6-plus
- Last updated: October 6, 2026

- Shared sourced benchmarks: 37
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Qwen3.6 Plus leads by point estimate. Conditional score ranges do not establish rank confidence. Choose using the category evidence and your own workload tests.

## Summary

- Qwen3.6 Plus has the higher overall point estimate 53.72 to 53.65. Conditional score ranges do not establish rank confidence.
- The clearest category separation is in instruction following, where the averages are 0 for Qwen3.5 397B and 82.5 for Qwen3.6 Plus.
- The biggest single benchmark swing is Terminal-Bench 2.0 in Agentic, with scores of 52.5% and 61.6%.
- Qwen3.6 Plus is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- Qwen3.6 Plus also has the larger context window at 1M.

## Model Snapshot

| Property | Qwen3.5 397B | Qwen3.6 Plus |
|----------|----------|----------|
| Creator | Alibaba | Alibaba |
| Type | Open Weight | Proprietary |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 128K | 1M |
| Overall Score | 53.65 | 53.72 |
| Benchmarks Covered | 38 | 48 |
| Pricing (input/output) | $0.60 / $3.60 | Pricing unavailable / Pricing unavailable |

## Category Breakdown

### Agentic

- Winner: Directional only
- Qwen3.5 397B public-lane score: 38.8 (Estimated · #62/120)
- Qwen3.6 Plus public-lane score: 33.8 (Supported · #75/120)

| Benchmark | Qwen3.5 397B | Qwen3.6 Plus | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 52.5% | 61.6% | Qwen3.6 Plus |
| BrowseComp | 62% | Coming soon | Coming soon |
| Claw-Eval | 56.8% | 58.8% | Qwen3.6 Plus |
| QwenClawBench | 51.8% | 57.2% | Qwen3.6 Plus |
| τ³-bench results | 68.4% | 70.7% | Qwen3.6 Plus |
| VITA-Bench | 43.7% | 44.3% | Qwen3.6 Plus |
| DeepPlanning | 37.6% | 41.5% | Qwen3.6 Plus |
| Toolathlon | 36.3% | 39.8% | Qwen3.6 Plus |
| MCP Atlas | 46.1% | 48.2% | Qwen3.6 Plus |
| MCP-Tasks | 74.2% | 74.1% | Qwen3.5 397B |
| WideResearch | 74.0% | 74.3% | Qwen3.6 Plus |
| Gert Labs | 46.76% | 50.60% | Qwen3.6 Plus |
| ResearchClawBench | 14.2% | 18.0% | Qwen3.6 Plus |
| Terminal-Bench 2.1 (Vals) | Coming soon | 53.2% | Coming soon |

### Coding

- Winner: Not comparable
- Qwen3.5 397B public-lane score: Coming soon
- Qwen3.6 Plus public-lane score: 41.7 (Supported · #59/144)

| Benchmark | Qwen3.5 397B | Qwen3.6 Plus | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 76.2% | 78.8% | Qwen3.6 Plus |
| LiveCodeBench v6 | 83.6% | 87.1% | Qwen3.6 Plus |
| SWE-bench Pro | 50.9% | 56.6% | Qwen3.6 Plus |
| SWE Multilingual | Coming soon | 73.8% | Coming soon |
| Vibe Code Bench | Coming soon | 25.56% | Coming soon |
| LiveCodeBench (Vals) | Coming soon | 86.0% | Coming soon |
| SWE-bench (Vals) | Coming soon | 73.4% | Coming soon |

### Reasoning

- Winner: Not comparable
- Qwen3.5 397B public-lane score: 60.9 (Unranked · 2 rankable rows)
- Qwen3.6 Plus public-lane score: 61.3 (Unranked · 4 rankable rows)

| Benchmark | Qwen3.5 397B | Qwen3.6 Plus | Winner |
|-----------|-----------|-----------|--------|
| LongBench v2 | 63.2% | 62% | Qwen3.5 397B |
| AI-Needle | 68.7% | 68.3% | Qwen3.5 397B |

### Multimodal & Grounded

- Winner: Qwen3.6 Plus
- Qwen3.5 397B public-lane score: 63.9 (#27/49)
- Qwen3.6 Plus public-lane score: 67.2 (#23/49)

| Benchmark | Qwen3.5 397B | Qwen3.6 Plus | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 79% | 78.8% | Qwen3.5 397B |
| MathVision | 88.6% | 88.0% | Qwen3.5 397B |
| CharXiv | 80.8% | 81.5% | Qwen3.6 Plus |
| VideoMMMU | 84.7% | 84.0% | Qwen3.5 397B |
| ScreenSpot Pro | 65.6% | 68.2% | Qwen3.6 Plus |
| V* | 95.8% | 96.9% | Qwen3.6 Plus |
| MMMU | Coming soon | 86.0% | Coming soon |

### Knowledge

- Winner: Directional only
- Qwen3.5 397B public-lane score: 49.1 (Estimated · #71/173)
- Qwen3.6 Plus public-lane score: 52.5 (Supported · #62/173)

| Benchmark | Qwen3.5 397B | Qwen3.6 Plus | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 88.4% | 90.4% | Qwen3.6 Plus |
| SuperGPQA | 70.4% | 71.6% | Qwen3.6 Plus |
| MMLU-Pro | 87.8% | 88.5% | Qwen3.6 Plus |
| MMLU-Redux | 94.9% | 94.5% | Qwen3.5 397B |
| C-Eval | 93% | 93.3% | Qwen3.6 Plus |
| HLE | 28.7% | 28.8% | Qwen3.6 Plus |
| GPQA Diamond (Vals) | Coming soon | 87.4% | Coming soon |
| MMLU-Pro (Vals) | Coming soon | 87.7% | Coming soon |

### Multilingual

- Winner: Tie
- Qwen3.5 397B public-lane score: 93.8 (#5/16)
- Qwen3.6 Plus public-lane score: 93.8 (#4/16)

| Benchmark | Qwen3.5 397B | Qwen3.6 Plus | Winner |
|-----------|-----------|-----------|--------|
| MMLU-ProX | 84.7% | 84.7% | Tie |
| NOVA-63 | 59.1% | 57.9% | Qwen3.5 397B |

### Instruction Following

- Winner: Directional only
- Qwen3.5 397B public-lane score: 0 (#125/125)
- Qwen3.6 Plus public-lane score: 82.5 (#48/125)

| Benchmark | Qwen3.5 397B | Qwen3.6 Plus | Winner |
|-----------|-----------|-----------|--------|
| IFEval | 92.6% | 94.3% | Qwen3.6 Plus |
| IFBench | Coming soon | 75.8% | Coming soon |

### Mathematics

- Winner: Not comparable
- Qwen3.5 397B public-lane score: 73.5 (Unranked · 5 rankable rows)
- Qwen3.6 Plus public-lane score: 62 (#5/7)

| Benchmark | Qwen3.5 397B | Qwen3.6 Plus | Winner |
|-----------|-----------|-----------|--------|
| AIME26 | 93.3% | 95.3% | Qwen3.6 Plus |
| HMMT Feb 2025 | 94.8% | 96.7% | Qwen3.6 Plus |
| HMMT Nov 2025 | 92.7% | 94.6% | Qwen3.6 Plus |
| HMMT Feb 2026 | 87.9% | 87.8% | Qwen3.5 397B |
| MMAnswerBench | 80.9% | 83.8% | Qwen3.6 Plus |
| FrontierMath v2 (Tiers 1-3) | Coming soon | 26.207% | Coming soon |
| FrontierMath v2 (Tier 4) | Coming soon | 8.333% | Coming soon |

## FAQ

### Which is better overall, Qwen3.5 397B or Qwen3.6 Plus?

Qwen3.6 Plus has the higher overall point estimate. Conditional score ranges do not establish rank confidence.

### Where is the biggest gap between Qwen3.5 397B and Qwen3.6 Plus?

The widest category gap is in instruction following, where the averages are 0 for Qwen3.5 397B and 82.5 for Qwen3.6 Plus.

### How many benchmarks does Qwen3.5 397B cover on BenchLM?

Qwen3.5 397B currently has 38 sourced benchmark scores on BenchLM.

### How many benchmarks does Qwen3.6 Plus cover on BenchLM?

Qwen3.6 Plus currently has 48 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Qwen3.5 397B vs Qwen3.5 397B (Reasoning)](/compare/qwen3-5-397b-vs-qwen3-5-397b-reasoning)
- [Qwen3.6 Plus vs Qwen3.5 397B (Reasoning)](/compare/qwen3-5-397b-reasoning-vs-qwen3-6-plus)
- [Qwen3.5 397B vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-qwen3-5-397b)
- [Qwen3.6 Plus vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-qwen3-6-plus)

## Explore More

- [Qwen3.5 397B profile](/models/qwen3-5-397b)
- [Qwen3.6 Plus profile](/models/qwen3-6-plus)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)